A statistical approach to rule selection in semantic query optimisation

Abstract

Abstract. Semantic Query Optimisation makes use of the semantic knowledge of a database (rules) to perform query transformation. Rules are normally learned from former queries fired by the user. Over time, however, this can result in the rule set becoming very large thereby degrading the efficiency of the system as a whole. Such a problem is known as the utility problem. This paper seeks to provide a solution to the utility problem through the use of statistical techniques in selecting and maintaining an optimal rule set. Statistical methods have, in fact, been used widely in the field of Knowledge Discovery to identify and measure relationships between attributes. Here we extend the approach to Semantic Query Optimisation using the Chi-square statistical method which is integrated into a prototype query optimiser developed by the authors. We also present a new technique for calculating Chi-square, which is faster and more efficient than the traditional method in this situation.

Extended Query Answering Using Integrity Rules

“…The process of evaluating this changed state against the integrity rules is then described in section 4 while finally our conclusions are presented in section 5. This work is an extension to the programme of research, by the authors, into the generation, evaluation and application of rules in databases [6,7,8,11].…”

A Fast Method for Ensuring the Consistency of Integrity Constraints

“…Thus recent developments in query optimisation [1,9,10,11,14] make use of the semantic information inherent in integrity constraints, together with rules derived from the data itself, to transform user queries into semantically equivalent alternatives which execute in far less time. Also many systems, such as taxation and social security, now incorporate facilities for interrogating the constraint base to provide answers to hypothetical 'what if' questions, e.g.…”

Study of authentic and fictitious online reviews

“…For the purpose of this study, IG and χ 2 were used to filter the set of 83 variables, and identify those that are important to distinguish between authentic and fictitious reviews. Variables with non-zero IG, and nonzero χ 2 values are generally deemed important for classification (Ananthakrishnan et al, 2011;Kumar & Valsala, 2013;Lowden & Robinson, 1999).…”

“…Variables with non-zero IG, and non-zero χ 2 values are generally deemed as important features for classification (Ananthakrishnan et al, 2011;Kumar & Valsala, 2013;Lowden & Robinson, 1999). In particular, 41 of the 83 variables had non-zero IG values as well as non-zero χ 2 values as shown in Table 4.6.…”